Monitoring individual fall risk using a wristband-type wearable sensor with a training-free method for proactive fall prevention

Falls are a major health concern, causing more than 10.9 million emergency room visits each year in the United States. Falls often result in severe injuries, such as hip fractures and traumatic brain injuries, which limit everyday activities such as dressing. Despite the seriousness, there is currently a lack of means to proactively prevent falls. In particular, our interviews with older adults, individuals with mobility issues, and those with a history of falls revealed an essential need for proactive measures. Caregivers also consistently expressed concern about their parents’ safety and frustration with the lack of preventive solutions. To address this unmet medical need, we developed a training-free machine learning model that can be incorporated into commercially available smartwatches and fitness bands to monitor an individual’s fall risk. Test results from 70 participants’ wearable data demonstrated that our method can effectively assess fall risk, achieving an average recall of 88.8%.

Discuss the importance of monitoring an individual’s fall risk to fill the unmet need in current practices that rely on reactive measures, such as connecting a fallen person to emergency services after an incident.

Examine whether a commercially available wristband sensor can be used to detect significant fall risk

Discuss the effectiveness of integrating physical and physiological data in a smartwatch to improve fall risk detection and reduce false alarms caused by fall-like activities (e.g., bending and jumping).